We have recently identified a type of neural activity that reflects the subject's engagement in a task. This task- related activity, which we have measured using both intrinsic-signal optical imaging and functional magnetic resonance imaging, is independent of external visual stimuli and is, instead, linked to internal brain states such as arousal and endogenous attention. Here, we propose a series of experiments that test several hypotheses about task-related brain activity, using a wide range of neurophysiological methods. Aim 1 will test whether hemodynamic activity (measured with optical imaging or with fMRI) can be separated into a linear sum of neurally distinct stimulus-evoked and task-related components. Aim 2 will test whether the task-related component is widespread but possibly a function of task modality (e.g. visual vs. auditory) and not spatially global. Aim 3 will test whether the task-related component reflects the degree of engagement (a form of arousal or attention) and is thus a function of reward, difficulty and timing. The endogenous processes that we propose to investigate are implicated a variety of neurological and psychiatric disorders. For example, individuals with ADHD demonstrate a wide variety of attentional deficits, but their most pronounced deficits are related to sustained attention, i.e., maintaining a steady level of performance on an infrequent task over a prolonged period of time. It is likely that the task-related component that we are proposing to study is related to sustained attention, and knowledge gained from the proposed experiments may provide insights into ADHD and related disorders.